Mists of Pandaria: All Specs - Page 29

Kaljurei · 02/11/13, 4:59 AM

I've seen a lot of separate posts from you guys on Arms in the PTR Thread, but it would help if someone posted a summary of the issues we're facing so we get a proper response to exactly that post, and other insightful ones.

So, just add or correct me if I'm wrong.

->1s GCD on Overpower makes the rotation clunky because we have to either wait 1s for MS, or clip MS by 0.5s.
->1s GCD on Overpower makes Sudden Death clunky because of the delay in the proc being affected by latency, an SD proc on the first of two Overpowers is not visible until we start pressing the second Overpower. Using Rage on a Slam, or using a Rage free filler in between doesn't help the mechanic because we could have used them inside a CS instead.
->Slam is still too low in priority compared to almost everything else, even with the Rage cost reduction and damage boost.
->Rage generation feels fine, but there are moments of rage starvation yet again even with a new rotation style, mostly due to the 10 rage cost on Overpower.
->Scaling is still abysmal, the Deep Wounds change has bumped Str scaling, but still no word on Haste and/or Mastery. We don't like numerical changes, but if that's the best you can come up with then so be it(May be too insulting to say, but that's that, since one of you US guys are going to post it, you can edit it if you want).
->Spec doesn't feel fun at all, because of bad rage generation, 1s GCD on Overpower making the rotation clunky, not great damage, etc.

Possible solutions :

->Keep the 1s GCD on Overpower, remove Sudden Death from Overpower and tie it to Autoattacks and Strikes of Opportunity at a chance between 10-20%, and probably decrease the Rage cost on Overpower to 5. Maybe introduce a mechanic that using Overpower reduces the Rage cost of the next Slam(can stack twice), but keep the 10 Rage cost on Overpower. Decrease MS's CD to 5s, or make Slam reduce the CD of MS by 1s or increase MS's CD by 0.5s, and increase the rage generated to 15/20.

What this does? We would only use Overpower when we want to save Rage for a CS burn phase. Reducing the rage cost of Slam would increase it's DPR. Haste and Mastery scaling are increased. We do lose a bit of control on Sudden Death, but the rotation feels a lot more fun to play with and the DPS loss should be offset by Haste and Mastery scaling and rage conservation.

Slam would also be used a lot more if it's DPR were increased. As from Berthold's post, it's DPR is worse than IV which is kind of pathetic.

->Rage generation could be slightly tweaked by bringing back Endless Rage(25% extra rage generation from autoattacks), or bring it back but let it apply to all Rage generation, i.e MS, Enrage, autoattacks etc, with a decreased modifier perhaps? We could have Strikes of Opportunity have generate rage instead, but a minute amount so that it's not too Overpowered.

What this does? We get more rage generation which would prevent rage starvation due to the 10s Rage cost on Overpower, maybe we use more HS, vastly helps in AOE scenarios to keep a fluid 2-target rotation or multitarget rotation. Increases Crit and probably Haste scaling due Enrage and Autoattacks generating more Rage. Increases Mastery scaling if SoO generates Rage.

Not sure what else I can suggest. One of the easiest ways to bump Crit scaling for Arms is bringing back Impale(20% extra Crit damage on MS, Slam and OP). Str scaling can be buffed by either buffing Deep Wounds(which has been done), buffing Execute or making Execute usable before 20% phase on a chance to proc(Which I'd may not have because it would compete with Slam then).

Seerow · 02/11/13, 11:10 AM

So, just add or correct me if I'm wrong.

->1s GCD on Overpower makes the rotation clunky because we have to either wait 1s for MS, or clip MS by 0.5s.

There is a third option: Go every third cycle without using TFB, and always use TFB in groups of 3. This gives us no wait time and no clipping of MS at all, and works really well. The problem comes from clipping of CS instead. Unlink CS from OP and it would actually work fairly well. Still not the most interesting rotation, but better than MS->OP->OP->X ad infinum or MS->OP->OP->X*2 ad infinum.

->Slam is still too low in priority compared to almost everything else, even with the Rage cost reduction and damage boost.

Slam's priority placement is hard. Simply push it above OP for example, turns the rotation into MS->Slamx3, which is even worse than what we have on PTR. So they have to be cautious about where to go with Slam's priority. Opening up more space in the rotation (which is what they tried to do with the 1s OP gcd) is the right direction.

->Scaling is still abysmal, the Deep Wounds change has bumped Str scaling, but still no word on Haste and/or Mastery. We don't like numerical changes, but if that's the best you can come up with then so be it(May be too insulting to say, but that's that, since one of you US guys are going to post it, you can edit it if you want).

My napkin math says that the deep wounds change makes for a difference of about .17 on strength scaling, leaving it still just a bit shy of 3.0. Prioritizing IV may improve it some as well though (since IV scales with AP rather than weapon damage). I also seem to remember seeing that IV deals 20% extra damage for Arms vs the other specs, but can't seem to find it again. Anyone have confirmation of that?

As for haste/mastery, reverting CS to auto attacks would bring haste back to where it was (roughly .9), then with the 50% buff should bring it around 1.35, which is still at the low end, but at least in the tolerable range.

->Spec doesn't feel fun at all, because of bad rage generation, 1s GCD on Overpower making the rotation clunky, not great damage, etc.

Adding on: It's not even that the rotation feels clunky or the damage isn't there. Ignoring scaling for a minute, as per Berthold the changes so far put us in the middle of the pack around the 509 ilvl. The main problem in my eyes that the rotation is boring. Like I can play Arms and not run into much rage starvation (waiting maybe a second or two per minute), and not feel too awkward, but the problem is that it's still boring. I remember when Arms was the more interesting alternative to Fury, where there was more to manage and keep track of, while Fury was the simple "hit whatever's available" spec. That is pretty much gone now, Arms rotation is a shadow of what it was, and is much less interesting as a result. What I want more than anything, even more than improved scaling, or competitive damage, is something extra to do.

GC made it clear on twitter that an extra proc is out of the question, they feel like the procs are what makes Fury feel furious, and are comfortable with Arms having only one proc. So instead they are trying to focus us on resource management, but the resource management aspect simply isn't enough as it stands to be truly compelling. It needs more meat to it to make it a interesting rotation. On that note:

->Rage generation could be slightly tweaked by bringing back Endless Rage(25% extra rage generation from autoattacks), or bring it back but let it apply to all Rage generation, i.e MS, Enrage, autoattacks etc, with a decreased modifier perhaps? We could have Strikes of Opportunity have generate rage instead, but a minute amount so that it's not too Overpowered.

This is something I definitely do not want. Well, extra rage might not be bad, but costs need to be balanced around it. Heroic Strike goes against everything a rage management focused rotation stands for, because it is so rage inefficient, that if we ever use it, it implies there is a failure. I don't mind it existing, and being there to be a point of failure (ie indicative of bad play) if you are forced to use it, but if the rotation itself is rage capped at any point that we consider using HS our only way to burn rage while playing optimally, there's something wrong with the rotation, because at that point the rotation is not effectively resource based. If that is how managing rage is going to work, then what we should see is another resource in play to manage.

Possible Alternative Suggestions I would make rotationally (not necessarily intended to be taken with each other):
->1s GCD for Arms on everything. Increase rage gen from auto attacks by 15% (brings it to slightly over 4.0 per second), remove rage generation from enrage (these balance out to roughly the same amount of rage overall, but more of it concentrated on auto attacks). CS procs from Auto Attacks and Mastery Strikes, Mastery Strikes proc half the rage of a normal auto attack, CS gains a rage cost of 10. Slam reduces MS cd by .5s per use.

This ends up being a net gain in rage gain, but the 1s gcd opens up a lot more room in the rotation, increasing our maximum rage expenditure greatly. Increasing MS usage helps boost Slam's effective damage per rage, and means as you have the haste to slam more, you also get more MSes (and thus more rage from MS). CS gaining a rage cost to make things feel more normal, fitting a rage based rotation better, and offsetting some of the extra rage gain from mastery procs.

Basically with this model our rotation has more wait time, but both Haste and Mastery scaling are much higher. It is a move more towards the Rogue/Feral/Monk model of resource management.

->Taste for Blood becomes a more managed resource. In addition to being able to be spent on OP, it can also be spent on one or two other things. Examples may be using Slam while having a TFB stack available consumes one stack and increases the crit chance on Slam, or a new ability you can use that provides a short duration damage buff but consumes all TFB stacks. For this one, CS would need to be unlinked from OP, either put onto Mastery/Attacks as above, or tied directly to usage of the TFB buff. Probably want to remove the reduced gcd as well. Here the cramped space for abilities is taken advantage of by forcing you to manage it efficiently.

In this scenario, TFB acts much as a secondary resource, it is normally used on OP (a very rage efficient attack), but depending on how many stacks you have and how much rage you have, you might choose to instead use your stacks for other things. Basically as you get more rage, you dump the stacks on other things. As long as the TFB abilities are designed well, this should boost haste scaling by virtue of letting us take advantage of increased rage in the rotation without losing efficiency and while still not having any wait time.

->Give Rend back, with a short duration (like 6-12 seconds, probably 9), using Slam extends the duration of Rend. Then make it so that you have either HS, or some new ability, can be used to consume all but one tick of Rend and deal the damage instantly to the target. This gives Arms a new high damage PVE nuke, something different/unique to manage, and doesn't really factor too much into PVP because you won't be able to stack a very high duration rend onto an enemy trying to run away from you. This solution works best with a mastery redesign that is trying to push Arms more towards bleeds, as has been suggested several times.

Aedilhild · 02/11/13, 12:08 PM

A super-concise list Arms' most objectively identified issues:

1) Overall DPS is low
2) Mastery/Haste scaling is poor
3) Necessary changes to limit Defensive Stance in PvP have resulted in low rage in Battle Stance, affecting AoE and utility
4) Priorities seem limited to a trifecta of CS, MS and Overpower (to the detriment of Slam and Heroic Strike)
5) Mechanics encouraging use of Sudden Death activator (Overpower) are counterproductive

Berthold · 02/12/13, 1:31 AM

I like aedilhilds list :-) btw there is a new simcraft release with the latest ptr data, inclduing the 1.2 buff to some spells for arms only. It would be great if some could hop on to the ptr and verify that the DPE are correct for a t14H equipped warrior. (or you could just take a look at how much more a strike hits than normal whitedmg, that should give a good approximation). Thanks and have fun testing the Spec / Priority changes. If you come up with something smarter, feel free to discuss here, or create a ticket at [1] so I can update the default list.

1: https://code.google.com/p/simulationcraft/issues/list

Berthold · 02/12/13, 1:34 AM

Another PTR update: OP reduces MS CD by 0.5. will do the coding afterwards, but apart from an even tighter rotation I don't expect any better scaling from this. Will most likely give us more MS > More Rage + less Slam -> More HS.

edit: results as expected. Here is the sim result: Arms 12.02.2013

p.s. If somebody comes up with an approximate BiS List for Arms/ Fury T15N/T15H, I could add it to simcraft. Chardev links would be great!

Kaljurei · 02/12/13, 3:34 AM

I'd really wish they had used Slam like Aedihild said instead of Overpower. Would have made so much more sense mechanically.

However, could we also ask that SD be removed from Overpower and be placed on either MS&Slam or Autoattacks&SoO? I think this could be mechanically more satisfying.

Kaljurei · 02/12/13, 4:36 AM

Originally Posted by Seerow

There is a third option: Go every third cycle without using TFB, and always use TFB in groups of 3. This gives us no wait time and no clipping of MS at all, and works really well. The problem comes from clipping of CS instead. Unlink CS from OP and it would actually work fairly well. Still not the most interesting rotation, but better than MS->OP->OP->X ad infinum or MS->OP->OP->X*2 ad infinum.

Yes, which is why unlinking CS from OP, but keeping the recent MS change would work extremely well.

Originally Posted by Seerow

Slam's priority placement is hard. Simply push it above OP for example, turns the rotation into MS->Slamx3, which is even worse than what we have on PTR. So they have to be cautious about where to go with Slam's priority. Opening up more space in the rotation (which is what they tried to do with the 1s OP gcd) is the right direction.

Well we cannot do MS->Slamx3 very frequently, as it would promote a lot of burst in PvP, which is what they're trying to avoid by putting a lot of advantages on using Overpower as well as keeping a rage cost of 10 to keep us from using the aforementioned combo frequently.

Originally Posted by Seerow

My napkin math says that the deep wounds change makes for a difference of about .17 on strength scaling, leaving it still just a bit shy of 3.0. Prioritizing IV may improve it some as well though (since IV scales with AP rather than weapon damage). I also seem to remember seeing that IV deals 20% extra damage for Arms vs the other specs, but can't seem to find it again. Anyone have confirmation of that?

Str scaling seems to have increased quite well in the new SimC release. I'm quite happy with this change.

As for IV's damage buff, here you go : MMO-Champion - Patch 5.2 - PTR Build 16539

IV's damage was increased for Arms from (1246 + 56% of AP) to (1495 + 67.2% of AP).

Originally Posted by Seerow

As for haste/mastery, reverting CS to auto attacks would bring haste back to where it was (roughly .9), then with the 50% buff should bring it around 1.35, which is still at the low end, but at least in the tolerable range.

Agreed. But frankly, I don't see a mechanical advantage to reforging for Haste or Mastery. If you look at other specs there's a clear advantage in most of the cases, and if not the other secondary stats scale quite well so it's compensated.

Haste would simply increase autoattack speed, which is really minimal even for a 10 minute fight as it results in only a few more autoattacks. Mastery on the other hand would allow us to do more damage per attack, which feels a lot more satisfying. If CS could be procced by both AA and SoO, then it would make a lot more sense, but Haste's value should still be low. I don't think we'd like Haste as a stat stronger than Mastery tbh.

Originally Posted by Seerow

Adding on: It's not even that the rotation feels clunky or the damage isn't there. Ignoring scaling for a minute, as per Berthold the changes so far put us in the middle of the pack around the 509 ilvl. The main problem in my eyes that the rotation is boring. Like I can play Arms and not run into much rage starvation (waiting maybe a second or two per minute), and not feel too awkward, but the problem is that it's still boring. I remember when Arms was the more interesting alternative to Fury, where there was more to manage and keep track of, while Fury was the simple "hit whatever's available" spec. That is pretty much gone now, Arms rotation is a shadow of what it was, and is much less interesting as a result. What I want more than anything, even more than improved scaling, or competitive damage, is something extra to do.

Yep. With the current change to Overpower reducing MS's CD, the rotation is pretty much going to look like :

MS->2x OP->Slam/CS/Execute/Heroic Throw/Storm Bolt/IV/DR. The problem lies in the last 2 GCDs, there are so many skills that can be used that Slam simply takes a lower priority than all of them except Heroic Throw. We could always disregard Execute because it happens only in 20% of the fight, but there are still too many alternatives. We can disregard SB in favor of BB also, again there are a lot many alternatives.

Originally Posted by Seerow

GC made it clear on twitter that an extra proc is out of the question, they feel like the procs are what makes Fury feel furious, and are comfortable with Arms having only one proc. So instead they are trying to focus us on resource management, but the resource management aspect simply isn't enough as it stands to be truly compelling. It needs more meat to it to make it a interesting rotation. On that note:

->Rage generation could be slightly tweaked by bringing back Endless Rage(25% extra rage generation from autoattacks), or bring it back but let it apply to all Rage generation, i.e MS, Enrage, autoattacks etc, with a decreased modifier perhaps? We could have Strikes of Opportunity have generate rage instead, but a minute amount so that it's not too Overpowered.

This is something I definitely do not want. Well, extra rage might not be bad, but costs need to be balanced around it. Heroic Strike goes against everything a rage management focused rotation stands for, because it is so rage inefficient, that if we ever use it, it implies there is a failure. I don't mind it existing, and being there to be a point of failure (ie indicative of bad play) if you are forced to use it, but if the rotation itself is rage capped at any point that we consider using HS our only way to burn rage while playing optimally, there's something wrong with the rotation, because at that point the rotation is not effectively resource based. If that is how managing rage is going to work, then what we should see is another resource in play to manage.

Well that's a good way to look at it. If consistent damage dealt and rage management are our priority, then I think we're already set. We perhaps need either a slight decrease in Rage generation because we're already using more Heroic Strikes than Slams(in Berthold's latest release), or we somehow make Slam more favorable. I don't see how we could do this unless we tie in SD to Slam, or make Slam do a crapload of damage but also cost more Rage.

Originally Posted by Seerow

Possible Alternative Suggestions I would make rotationally (not necessarily intended to be taken with each other):
->1s GCD for Arms on everything. Increase rage gen from auto attacks by 15% (brings it to slightly over 4.0 per second), remove rage generation from enrage (these balance out to roughly the same amount of rage overall, but more of it concentrated on auto attacks). CS procs from Auto Attacks and Mastery Strikes, Mastery Strikes proc half the rage of a normal auto attack, CS gains a rage cost of 10. Slam reduces MS cd by .5s per use.

I like it, but I'd rather we kept the rage generation from Enrage simply because it's more of a guaranteed rage increase when you don't have enough uptime on a target(i.e PvP). Also, Crit scaling goes down and Haste scaling goes up, I think this is exactly what we don't want, we'd rather have only Mastery scaling go up and Haste scale a lot lesser so Mastery's the second secondary stat we reforge into, rather than Haste.

Considering a 1s GCD, We have 5 GCDs to use between 2 Mortal Strikes. So that would be CS>Slamx2>OP all the time, but do we have enough Rage generation to use Slamx2 every MS cycle, on top of a 10 Rage cost on CS? I'd need a mathematical model to support this theory. It's 60 Rage spent every MS cycle, I don't think there's enough even with all of the changes you described.

Originally Posted by Seerow

->Taste for Blood becomes a more managed resource. In addition to being able to be spent on OP, it can also be spent on one or two other things. Examples may be using Slam while having a TFB stack available consumes one stack and increases the crit chance on Slam, or a new ability you can use that provides a short duration damage buff but consumes all TFB stacks. For this one, CS would need to be unlinked from OP, either put onto Mastery/Attacks as above, or tied directly to usage of the TFB buff. Probably want to remove the reduced gcd as well. Here the cramped space for abilities is taken advantage of by forcing you to manage it efficiently.

In this scenario, TFB acts much as a secondary resource, it is normally used on OP (a very rage efficient attack), but depending on how many stacks you have and how much rage you have, you might choose to instead use your stacks for other things. Basically as you get more rage, you dump the stacks on other things. As long as the TFB abilities are designed well, this should boost haste scaling by virtue of letting us take advantage of increased rage in the rotation without losing efficiency and while still not having any wait time.

I proposed a very similar idea where MS simply granted 2 charges of Overpower, and using Overpower granted you a buff called Taste for Blood, which could be consumed to increase damage by X% for Y Seconds, by using another off-GCD ability which could be called Bloodrage(Just a placeholder for a name, we did have an ability like this before, and it seemed appropriate).

I think however, they're already moving in the right direction by making OP reduce MS's CD, and proposing a damage buff at this point with GC's latest post in context, may seem like overkill.

Could we also have any kind of suggestions on the Set Bonuses?

2pc : I guess it's fine for us except the proc chance isn't great and it simply provides a passive damage boost every now and then. But it doesn't feel fun. It feels fun for Fury because they get a Raging Blow proc but for Arms it's just...boring.

4pc : I don't like this set bonus at all. Skull Banner is a Raid Offensive cooldown and should be used as such. I understand that it's a kind of compensation for us that we use a Raid CD and get a bigger benefit out of it, but it really screws things over when there are more Warriors in the raid, and pairing Reck with SB. I think this needs to be looked at urgently.

Seerow · 02/12/13, 7:19 AM

Honestly based on that simcraft result, assuming it is accurate, I think Scaling is good enough to be left alone as it is. I'm surprised to see the crit scaling pick up more, I guess more MS usage means more enrage procs, which makes crit more useful. But between Crit going from average to above average, and then weapon dps and strength both picking up a bit (Strength scaling went up about 50%, weapon dps up about 10%), that pulls our overall primary scaling well ahead of the middle of the pack where it was, and the two combined are now enough to make our overall scaling fairly solid. I at least am placated for the time being. I think that secondary scaling is more satisfying, and would be something nice to look at in the long term (ie start clammoring for it around 6.0), but what is there will keep us competitive.

I'll respond to the other stuff in a couple of hours. Don't have time right this second.

Aedilhild · 02/12/13, 8:12 AM

Originally Posted by Kaljurei

I'd really wish they had used Slam like Aedihild said instead of Overpower. Would have made so much more sense mechanically.

However, could we also ask that SD be removed from Overpower and be placed on either MS&Slam or Autoattacks&SoO? I think this could be mechanically more satisfying.

Part of me wonders if it's a programming issue for Blizzard. The other part is amused that a player asks for Slam, and all Blizzard hears is "Overpower."

In any case, I'm as impressed with the Crit scaling in Berthold's sim as Secondwind is, since it was intended for QoL. (I imagine the same could be achieved with Slam, but someone needs to get Irvine on speed-dial.) As to what you're asking, I wonder if the "random" nature of prior Sudden Death was more the uncontrollable loss of Colossus Smash uses, especially when Mastery procced the ability. Maybe if Colossus Smash could hold two charges, Sudden Death could go back to 6% on all melee -- that might shore up Mastery and Haste, and work with Arms' "cool and calculating" theme. From that point, any issue with too-high damage (ironically) could be curbed by lowering Seasoned Soldier.

Seerow · 02/12/13, 9:59 AM

Part of me wonders if it's a programming issue for Blizzard. The other part is amused that a player asks for Slam, and all Blizzard hears is "Overpower."

Yeah, I see a lot of posts on the PTR forums saying the same thing. You make the suggestion for Slam, and they put it in on Overpower.

Ideally some combination of your suggestion and my second suggestion above would be implemented. Make Slam reduce the cooldown on MS and be a 1s gcd. Make CS proc from White/Mastery, or all strikes, as opposed to from Overpower, with up to two charges. Make sure OP stays more efficient than Slam, even with the changes (Slam giving more MS uses makes it more efficient than a pure damage/rage analysis would show, OP may need to be buffed or have its rage cost dropped to make sure OP stays a priority when possible), and then give us a new ability to funnel excess TFB stacks into.

Overall this would lead to us have an always-GCD capped playstyle, that at low rage we use MS->OP->OP->X, but at high rage will use MS->Slamx3 to try to bleed off rage. Bleeding off rage that way causes a backlog of TFB stacks to build up, and if we get to the point where we have more than we can spend (sitting around 4-5), we dump them with the new ability and slam some more until rage is back at a manageable level. So the baseline rage cost for the rotation is roughly 60 per 11.5 seconds (ie 1 slam every 2 MS cycles), which is easily maintained. Maximum rage consumption is 60 per 4.5 seconds, which requires way more haste than we will ever have, ie leaving lots of room for the rotation to grow as we gear up, without ever feeling rage starved or like we can't press any buttons. Most of the time you'll be managing some mixture of the two, depending on CS procs and what other cooldowns are available.

As you noted, any increase in DPS resulting from this can be fixed by toning down Seasoned Soldier.

Agreed. But frankly, I don't see a mechanical advantage to reforging for Haste or Mastery. If you look at other specs there's a clear advantage in most of the cases, and if not the other secondary stats scale quite well so it's compensated.

Haste would simply increase autoattack speed, which is really minimal even for a 10 minute fight as it results in only a few more autoattacks. Mastery on the other hand would allow us to do more damage per attack, which feels a lot more satisfying. If CS could be procced by both AA and SoO, then it would make a lot more sense, but Haste's value should still be low. I don't think we'd like Haste as a stat stronger than Mastery tbh.

Honestly, Mastery isn't satisfying to me at all. It's not a 'fun' stat in the way that haste is for other specs, or crit is for fury. It's a passive second crit that hits so soft you generally don't even notice it.

I tend to lean heavily on things that help secondary scaling that also make those secondaries more interaction. For example with the above suggestion, more haste leads to being able to hit slam more which leads to dumping the TFB stacks on the other ability more. Basically more haste changes the way you play, which is the way it is for other classes, to greater or lesser degrees. Or mastery proccing bonus rage/enabling CS, again something that changes the way we play. Not all stats need to do that, but all specs should have at least one.

I like it, but I'd rather we kept the rage generation from Enrage simply because it's more of a guaranteed rage increase when you don't have enough uptime on a target(i.e PvP).

Honestly, lack of rage when missing uptime is an issue that needs to be dealt with separately. I discussed it here

Considering a 1s GCD, We have 5 GCDs to use between 2 Mortal Strikes. So that would be CS>Slamx2>OP all the time, but do we have enough Rage generation to use Slamx2 every MS cycle, on top of a 10 Rage cost on CS? I'd need a mathematical model to support this theory. It's 60 Rage spent every MS cycle, I don't think there's enough even with all of the changes you described.

That's the point. We wouldn't have enough rage to maintain full GCD lock at 1s gcd. The point is to make haste more valuable by having the rage it provides let us press more abilities. Like I said, that particular suggestion was aiming to make us scale more like other resource based classes who have some wait time and get to fill more globals as they get more resources. If you prefer having 100% gcd lock, something more along the lines of the second suggestion (or the modified version at the beginning of this post) would be the best bet.

Kaljurei · 02/12/13, 11:34 AM

Originally Posted by Aedilhild

Part of me wonders if it's a programming issue for Blizzard. The other part is amused that a player asks for Slam, and all Blizzard hears is "Overpower."

In any case, I'm as impressed with the Crit scaling in Berthold's sim as Secondwind is, since it was intended for QoL. (I imagine the same could be achieved with Slam, but someone needs to get Irvine on speed-dial.) As to what you're asking, I wonder if the "random" nature of prior Sudden Death was more the uncontrollable loss of Colossus Smash uses, especially when Mastery procced the ability. Maybe if Colossus Smash could hold two charges, Sudden Death could go back to 6% on all melee -- that might shore up Mastery and Haste, and work with Arms' "cool and calculating" theme. From that point, any issue with too-high damage (ironically) could be curbed by lowering Seasoned Soldier.

Well, SD is only controllable now because it's tied to an ability, instead of autoattacks. On the other hand, I believe that the Cata SD Mechanic provided bigger uptime of CS than the newer versions have.

I'd definitely support the charge mechanism, but the question there would also be why separate SD from Overpower when you can keep charges. I don't think at this point GC would support any scaling changes unless he was sure that the current projected DPS is lower than what should be. But of course like you said we could lower the damage by lowering Seasoned Soldier, so yeah it could definitely work. Personally, I like having charges on an ability, it makes it a lot more fun, it was an awesome design change when the concept of charges on Raging Blow was introduced, I don't see why Sudden Death should be any different as it plays a pretty similar role, in providing a lot of damage.

Originally Posted by Seerow

Honestly, Mastery isn't satisfying to me at all. It's not a 'fun' stat in the way that haste is for other specs, or crit is for fury. It's a passive second crit that hits so soft you generally don't even notice it.

I tend to lean heavily on things that help secondary scaling that also make those secondaries more interaction. For example with the above suggestion, more haste leads to being able to hit slam more which leads to dumping the TFB stacks on the other ability more. Basically more haste changes the way you play, which is the way it is for other classes, to greater or lesser degrees. Or mastery proccing bonus rage/enabling CS, again something that changes the way we play. Not all stats need to do that, but all specs should have at least one.

Well Haste has such a negligible effect on our Rage generation that your point doesn't make a lot of sense for Haste. Mastery proccing bonus Rage is an idea I like though.

Originally Posted by Seerow

Honestly, lack of rage when missing uptime is an issue that needs to be dealt with separately. I discussed it here

That is a nice idea.

Originally Posted by Seerow

That's the point. We wouldn't have enough rage to maintain full GCD lock at 1s gcd. The point is to make haste more valuable by having the rage it provides let us press more abilities. Like I said, that particular suggestion was aiming to make us scale more like other resource based classes who have some wait time and get to fill more globals as they get more resources. If you prefer having 100% gcd lock, something more along the lines of the second suggestion (or the modified version at the beginning of this post) would be the best bet.

Haste needs to do more than just increasing autoattack speed if you want it to have an effect on Rage generation. It needs to increase rage generated from MS and Enrage as well. I had advocated an idea like Sanctity of Battle where Haste provides a proportional increase to the amount of Rage generated by MS and Enrage, i.e X% of Haste causes a Y increase in the amount of rage generated from MS and Enrage.

Seerow · 02/12/13, 12:04 PM

Uh, I have a problem with Slam reducing MS's CD instead.

Considering Slam reduces MS's CD by 1 second :

0.0 MS
1.5 Slam - MS CD now at 3.5 - (6-1.5-1)
3.0 Slam - MS CD now at 1s (3.5-1.5-1)
4.5 MS (We delayed MS by 0.5s because it comes off CD at 4.0)

Does this look right or is my math wrong? Using 2x Slams with 2 GCDs makes MS go down to a 4s CD. How much damage did we do? 175%+(220%*2)=615% WD in 4s costing 40 Rage.

Comparing to current mechanic of Overpower reducing MS's CD by 0.5s

0.0 MS
1.5 OP - MS CD now at 4s (6-1.5-1)
2.5 OP - MS CD now at 2.5s (4-1-0.5)
3.5 Slam
5.0 MS

You missed an important part. Particularly:

Make Slam reduce the cooldown on MS and be a 1s gcd.

Basically, taking the two newest toys from OP, and putting them on Slam instead.

So your low rage rotation is:

0s-MS
1.5s-OP
3.0s-OP
4.5s-Cheap/Free Filler
6s-MS

Your high rage rotation is:
0s-MS
1.5s-Slam
2.5s-Slam
3.5s-Slam
4.5s-MS

No wait time involved.

Any combination of replacing OPs/Fillers with slams results in no delay.

I'm ignoring the math you posted, because while I'm sure it's right, it's in relation to a rotation that wasn't intended.

[Edit: Basically I see it working something like:
First set with a slam as the filler: 175%+105%+105%+220%

Add in ~20% crit rate, and 80% for OP and you get

212.1+194+194+266.6 = 866.7% weapon damage in 5.5 seconds for 40 rage. (7.2 rage per second)

Second set spamming Slam is

212.1+266.6+266.6+266.6 = 1011.9 damage in 4.5 seconds for 60 rage. (13.3 rage per second) (increase in rage consumption of 85%. The difference is even greater when you take into account other fillers instead of slam for the first rotation, ie Dragon's Roar/Impending Victory)

Also note this is without taking into account the intended damage buff from dumping TFB stacks into the buff rather than OPs, simply because I'm not sure what damage value it would need to be tuned at to be better than dumping rage into HS while continuing to use OP, but worse than ignoring OP altogether

Edit2: Also worth noting, if the filler is Impending Victory instead of Slam, you instead are spending 30 rage over 6 seconds, or 5 rage per second. If you use a free filler ie DR/CS, it's instead 20 rage per 6 seconds, or 3.3 rage per second. So depending on which abilities are available, rage consumption will vary from less than half of our generation to as much as double what we generate, forcing us to shake things up on the fly to keep using the abilities we need to while avoiding capping resources as much as possible]

Well Haste has such a negligible effect on our Rage generation that your point doesn't make a lot of sense for Haste. Mastery proccing bonus Rage is an idea I like though.

****

Haste needs to do more than just increasing autoattack speed if you want it to have an effect on Rage generation. It needs to increase rage generated from MS and Enrage as well. I had advocated an idea like Sanctity of Battle where Haste provides a proportional increase to the amount of Rage generated by MS and Enrage, i.e X% of Haste causes a Y increase in the amount of rage generated from MS and Enrage.

In the 1s gcd scenario, Haste would have a higher effect on rage generation. Remember, that suggested rotation I had buffed rage from auto attacks, and lost rage from enrage effects, so haste gives more rage from auto attacks. That more rage allows you to slam more, which then reduces the MS cooldown, thus effectively giving more rage. It's not neat and immediately observable, but the scaling is there. If you're bored and have the time, try programming it into Simcraft, see what comes up. I'd be interested to see it.

(I really need to learn how to program that thing so I can make that work myself. Is there a resource that shows how?)

As an aside, a sanctity of battle-esque ability is something I've wanted since before Sanctity of Battle existed. Something as simple as reducing MS cooldown and all GCDs by haste % would instantly bring our haste scaling up to normal levels, and make it a much more fun stat. Just recently I've been focusing instead on coming up with other potential ideas, because that is one the developers have made clear they don't intend to pursue.

Kaljurei · 02/12/13, 12:54 PM

Yeah my calculation was wrong, sorry I missed that part.

But again, do you think that your suggestion is sustainable? There's no way we're going to be able to generate 13.3 RPS without doubling the rage from autoattacks, or splitting it in between MS/Enrage and Mastery and Autoattacks. Also, I think it's a massive buff to be able to use Slamx3.

I honestly think your idea could be implemented, but I doubt we're going to enter such a paradigm shift in the gameplay for Arms in mid-expac.

Aedilhild · 02/12/13, 1:16 PM

Originally Posted by Kaljurei

Also, I think it's a massive buff to be able to use Slamx3.

Well, the original thought process for my counter-suggestion with Slam was:

1) A 1-sec. GCD spoiler introduces a regular delay that nostalgia probably doesn't justify
2) Reducing MS's CD by 0.5 sec. allows interlock for rotational attacks
3) Overpower already has quite a few incentives, plus the increased instance of overwriting Sudden Death
4) Slam doesn't have a lot going for it, and is free of special effects

The aim would not be for Arms players to go triple-Slam all the time. In fact, I'd expect its rage cost to moderate usage and still punish over-usage. The aim would be avoiding conflict with existing mechanics like Sudden Death, while balancing priorities a little more between Overpower and everything else. And that is even predicated on Blizzard's decision on whether Arms needs a quirk after all.

Seerow · 02/12/13, 1:17 PM

Originally Posted by Kaljurei

Yeah my calculation was wrong, sorry I missed that part.

But again, do you think that your suggestion is sustainable? There's no way we're going to be able to generate 13.3 RPS without doubling the rage from autoattacks, or splitting it in between MS/Enrage and Mastery and Autoattacks. Also, I think it's a massive buff to be able to use Slamx3.

I honestly think your idea could be implemented, but I doubt we're going to enter such a paradigm shift in the gameplay for Arms in mid-expac.

The rotation would be sustainable. The point isn't to make it so that you spam Slamx3 every cycle. In fact, the point is that you are never actually able to do that. If you are able to spam nothing but MS->Slamx3, there is a very real problem because then the rotation is even more boring than live. What you want is for there to be points in the rotation where you do so. Right now those points are where we hit heroic strike.

[edit: Aedilhild covered this point more succinctly than I did.]

Under the paradigm I describe, we'd instead change up our rotation to accommodate that extra rage. The whole point is to make Arms more resource focused, so rather than having a set priority you hit every time, you change how you play depending on what your resources are as the fight goes. You will generally play conservatively at the start of the fight, using a lot of your free abilities that are on cooldowns (ie CS, DR, IV), and the cheap Overpower. As the fight goes on, your rage bar is building up faster than you are spending it while you focus on the cheaper abilities, and you start shifting your focus to using more slams. In doing so, you start having a backlog of TFB stacks you aren't using. Eventually you hit a critical mass where you have 4-5 stacks of TFB you haven't used, and are high on rage, and at this point you dump the TFB stacks with the new ability, and use a bunch of slams, basically resetting you to the start.

The better question, as you point out, is whether or not there is a chance it would be implemented mid-expansion. But honestly, they've already gone so far in changing Arms, that something like this would be needed. If Arms goes live the way it is on the PTR, despite the great increases in damage, I will likely end up switching classes. Because Arms on the PTR lacks resource management, upkeep, and procs. You really need at least one of the three to keep things interesting. As it is, if it goes live the rotation is so mind numbingly simple that it will be nearly impossible to mess up, and that was the problem I had with Cata fury that got me to switch to Arms in the first place. It doesn't need to this exact suggestion (I've offered at least two others, and I've seen several others, including a couple from Aedilhild that I would be happy with as well), but a serious shift is needed, and it is frustrating to see Blizzard coming so close to making something great, then falling just short and going the opposite direction.